Counterfactual Reward Model Training for Bias Mitigation in Multimodal Reinforcement Learning

• URL: http://arxiv.org/abs/2508.19567v1
• Date: Wed, 27 Aug 2025 04:54:33 GMT
• Title: Counterfactual Reward Model Training for Bias Mitigation in Multimodal Reinforcement Learning
• Authors: Sheryl Mathew, N Harshit,
• Abstract summary: We present a counterfactual reward model that introduces causal inference with multimodal representation learning to provide an unsupervised, bias-resilient reward signal.<n>We evaluated the framework on a multimodal fake versus true news dataset, which exhibits framing bias, class imbalance, and distributional drift.<n>The resulting system achieved an accuracy of 89.12% in fake news detection, outperforming the baseline reward models.
• Score: 0.5204229323525671
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In reinforcement learning with human feedback (RLHF), reward models can efficiently learn and amplify latent biases within multimodal datasets, which can lead to imperfect policy optimization through flawed reward signals and decreased fairness. Bias mitigation studies have often applied passive constraints, which can fail under causal confounding. Here, we present a counterfactual reward model that introduces causal inference with multimodal representation learning to provide an unsupervised, bias-resilient reward signal. The heart of our contribution is the Counterfactual Trust Score, an aggregated score consisting of four components: (1) counterfactual shifts that decompose political framing bias from topical bias; (2) reconstruction uncertainty during counterfactual perturbations; (3) demonstrable violations of fairness rules for each protected attribute; and (4) temporal reward shifts aligned with dynamic trust measures. We evaluated the framework on a multimodal fake versus true news dataset, which exhibits framing bias, class imbalance, and distributional drift. Following methodologies similar to unsupervised drift detection from representation-based distances [1] and temporal robustness benchmarking in language models [2], we also inject synthetic bias across sequential batches to test robustness. The resulting system achieved an accuracy of 89.12% in fake news detection, outperforming the baseline reward models. More importantly, it reduced spurious correlations and unfair reinforcement signals. This pipeline outlines a robust and interpretable approach to fairness-aware RLHF, offering tunable bias reduction thresholds and increasing reliability in dynamic real-time policy making.

Related papers

• Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation [23.667292152163068]
  We present textbfDSRM-HRL, a framework that reformulates fairness-aware recommendation as a latent state purification problem.<n>We show that DSRM-HRL effectively breaks the "rich-get-richer" feedback loop, achieving a superior frontier between recommendation utility and exposure equity.
  arXiv  Detail & Related papers  (2026-03-04T08:14:21Z)
• Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling [49.41422138354821]
  We propose a principled reward modeling framework that integrates non-negative factor analysis into the Bradley-Terry preference model.<n>BNRM represents rewards through a sparse, non-negative latent factor generative process.<n>We show that BNRM substantially mitigates reward over-optimization, improves robustness under distribution shifts, and yields more interpretable reward decompositions than strong baselines.
  arXiv  Detail & Related papers  (2026-02-11T08:14:11Z)
• Mitigating Estimation Bias with Representation Learning in TD Error-Driven Regularization [4.784045060345404]
  This work introduces enhanced methods to achieve flexible bias control and stronger representation learning.<n>We propose three convex combination strategies, symmetric and asymmetric, that balance pessimistic estimates to mitigate overestimation and optimistic exploration via double actors.<n>To further improve performance, we integrate augmented state and action representations into the actor and critic networks.
  arXiv  Detail & Related papers  (2025-11-20T06:31:55Z)
• Drift No More? Context Equilibria in Multi-Turn LLM Interactions [58.69551510148673]
  contexts drift is the gradual divergence of a model's outputs from goal-consistent behavior across turns.<n>Unlike single-turn errors, drift unfolds temporally and is poorly captured by static evaluation metrics.<n>We show that multi-turn drift can be understood as a controllable equilibrium phenomenon rather than as inevitable decay.
  arXiv  Detail & Related papers  (2025-10-09T04:48:49Z)
• Learning a Dense Reasoning Reward Model from Expert Demonstration via Inverse Reinforcement Learning [50.20267980386502]
  We learn a dense, token-level reward model for process supervision directly from expert demonstrations.<n>The learned reasoning reward serves two complementary roles: (i) it provides step-level feedback to optimise a reasoning policy during training; and (ii) it functions at inference as a critic to rerank sampled traces under fixed compute budgets.
  arXiv  Detail & Related papers  (2025-10-02T09:55:26Z)
• Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment [30.605500809158986]
  We propose a novel causal reward modeling approach that integrates causality to mitigate spurious correlations.<n>Our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences.
  arXiv  Detail & Related papers  (2025-01-16T16:00:37Z)
• Fair CoVariance Neural Networks [34.68621550644667]
  We propose Fair coVariance Neural Networks (FVNNs), which perform graph convolutions on the covariance matrix for both fair and accurate predictions.<n>We prove that FVNNs are intrinsically fairer than analogous PCA approaches thanks to their stability in low sample regimes.
  arXiv  Detail & Related papers  (2024-09-13T06:24:18Z)
• Preference-Based Multi-Agent Reinforcement Learning: Data Coverage and Algorithmic Techniques [65.55451717632317]
  We study Preference-Based Multi-Agent Reinforcement Learning (PbMARL)<n>We identify the Nash equilibrium from a preference-only offline dataset in general-sum games.<n>Our findings underscore the multifaceted approach required for PbMARL.
  arXiv  Detail & Related papers  (2024-09-01T13:14:41Z)
• Perturbation-Invariant Adversarial Training for Neural Ranking Models: Improving the Effectiveness-Robustness Trade-Off [107.35833747750446]
  adversarial examples can be crafted by adding imperceptible perturbations to legitimate documents. This vulnerability raises significant concerns about their reliability and hinders the widespread deployment of NRMs. In this study, we establish theoretical guarantees regarding the effectiveness-robustness trade-off in NRMs.
  arXiv  Detail & Related papers  (2023-12-16T05:38:39Z)
• RobustFair: Adversarial Evaluation through Fairness Confusion Directed Gradient Search [8.278129731168127]
  Deep neural networks (DNNs) often face challenges due to their vulnerability to various adversarial perturbations. This paper introduces a novel approach, RobustFair, to evaluate the accurate fairness of DNNs when subjected to false or biased perturbations.
  arXiv  Detail & Related papers  (2023-05-18T12:07:29Z)
• Delving into Identify-Emphasize Paradigm for Combating Unknown Bias [52.76758938921129]
  We propose an effective bias-conflicting scoring method (ECS) to boost the identification accuracy. We also propose gradient alignment (GA) to balance the contributions of the mined bias-aligned and bias-conflicting samples. Experiments are conducted on multiple datasets in various settings, demonstrating that the proposed solution can mitigate the impact of unknown biases.
  arXiv  Detail & Related papers  (2023-02-22T14:50:24Z)
• Explicit Tradeoffs between Adversarial and Natural Distributional Robustness [48.44639585732391]
  In practice, models need to enjoy both types of robustness to ensure reliability. In this work, we show that in fact, explicit tradeoffs exist between adversarial and natural distributional robustness.
  arXiv  Detail & Related papers  (2022-09-15T19:58:01Z)
• Bias-inducing geometries: an exactly solvable data model with fairness implications [12.532003449620607]
  We introduce an exactly solvable high-dimensional model of data imbalance.<n>We analytically unpack the typical properties of learning models trained in this synthetic framework.<n>We obtain exact predictions for the observables that are commonly employed for fairness assessment.
  arXiv  Detail & Related papers  (2022-05-31T16:27:57Z)
• Adversarial Robustness under Long-Tailed Distribution [93.50792075460336]
  Adversarial robustness has attracted extensive studies recently by revealing the vulnerability and intrinsic characteristics of deep networks. In this work we investigate the adversarial vulnerability as well as defense under long-tailed distributions. We propose a clean yet effective framework, RoBal, which consists of two dedicated modules, a scale-invariant and data re-balancing.
  arXiv  Detail & Related papers  (2021-04-06T17:53:08Z)
• Trust but Verify: Assigning Prediction Credibility by Counterfactual Constrained Learning [123.3472310767721]
  Prediction credibility measures are fundamental in statistics and machine learning. These measures should account for the wide variety of models used in practice. The framework developed in this work expresses the credibility as a risk-fit trade-off.
  arXiv  Detail & Related papers  (2020-11-24T19:52:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.